Simple question

[AnalogKid]

The operating system was punched cards and the RAM memory was thousands of tiny donuts of ferrite hanging on a maze of wires.

I miss core memory. I have a Data General 16 K core board on my office wall as art. 30 years after it was pulled, if I plugged it into working chassis and hit RUN it would pick up on the next instruction Fizix.

ak

 

[granddad]

I worked ( as a field tech guy ) on a cash register with core memory , then they tried bubble , followed by battery backed cmos, and finally NVRAM .... all with strange voltages like -13v , most machines hated static ... I had to ask the operator they had nylon underwear !

 

[John Murdock]

Question:

In terms of current/voltage I know that each LED requires 2v, is this the same for each LED in a segment display?
I remeber reading about internal resistance of CMOS chips affecting this. In theory if there is 7 LEDS that's 14v mind ones that have a decimal point which is 16v?

I used 9v batteries but this works,how does the driver manage this and if so how?

 

[spec]

Question:

In terms of current/voltage I know that each LED requires 2v, is this the same for each LED in a segment display?
I remeber reading about internal resistance of CMOS chips affecting this. In theory if there is 7 LEDS that's 14v mind ones that have a decimal point which is 16v?

I used 9v batteries but this works,how does the driver manage this and if so how?

John,

Your attachment does not render.

spec

 

[John Murdock]

John,

Your attachment does not render.

spec

There isn't meant to be an attachment?

I was referring to same style schematic in post #1

 

[AnalogKid]

The 8 LEDs in a display are in parallel, not series, so the net voltage drop is the standard 1.8V-2.1V (depending on color, manufacturer, etc.). For an individual character display, the Common Cathode or Common Anode connection goes to all segments in the character, and the other end of each LED comes out on a separate pin. For multi-digit displays with internal multiplexing connections, things get a bit more complex. But in all cases, an individual driver circuit sees only one diode at a time.

ak

 

[John Murdock]

when doing a calculation for monostable time period (t=1.1*CR)
capacitor=100uf
resistor = 10k

so its T= 1.1 * (100* 10^-6) * (10* 10^-3)
T = 0.0000011 seconds seems wrong, very wrong? what's wrong.


I've done much harder calculations than this but this I mean am I converting them incorrectly?

 

[spec]

when doing a calculation for monostable time period (t=1.1*CR)
capacitor=100uf
resistor = 10k

so its T= 1.1 * (100* 10^-6) * (10* 10^-3)
T = 0.0000011 seconds seems wrong, very wrong? what's wrong.


I've done much harder calculations than this but this I mean am I converting them incorrectly?

Unless I have gone wrong the answer is 1.1 seconds. Your mistake is that you have made the 10K resistor 10 mili Ohms.

I did a similar thing the other day, but I was only three orders out, instead of six.

spec

 

[John Murdock]

Unless I have gone wrong the answer is 1.1 seconds. Your mistake is that you have made the 10K resistor 10 mili Ohms.

I did a similar thing the other day, but I was only three orders out, instead of six.

spec

Aha right that's it . Totally feel like an idiot right now.

 

[spec]

Aha right that's it . Totally feel like an idiot right now.

Not an idiot- gross errors like that are made all the time. I probably make a similar error a couple of times a day. It's the subtle errors that are the killers. Using e in place of pi is a good one.

Here is an example of a three order error: https://www.electro-tech-online.com/threads/milliohmeter.147743/#post-1259750
I felt a right dick with that post

On a serious note, an engineer, or anyone who does anything in this world, has to come to terms with errors. They are just a fact of life.

spec

 

[John Murdock]

One other thing, I was playing around with some hobby projects and I saw that segment g on a 7seg display wasn't lighting up when all the others were so as you do I turned the PCB around looked at the soldering, tracks etc it was perfect, could it be the display itself not working keep in mind that the display has rarely been used infact when I made it I really hadn't used it for anything.

 

[spec]

One other thing, I was playing around with some hobby projects and I saw that segment g on a 7seg display wasn't lighting up when all the others were so as you do I turned the PCB around looked at the soldering, tracks etc it was perfect, could it be the display itself not working keep in mind that the display has rarely been used infact when I made it I really hadn't used it for anything.

You can confirm if the g segment is faulty by measuring the volts across it. If it has 2V or so across it and it is not illuminating it is faulty, but unless it has been stressed, it is highly unlikely to be faulty. Apart from obvious abuse I have never known a faulty LED.

spec

 

[Tony Stewart]

Circuit: input - debouncer - 4026 - 7 seg

one could connect in other 4026b chips in order to show multiple digits e.g 2 chips up to 99 3 chips 999 4 chips 9999 Etc.

This could be done by connecting pin 5 of the first IC and to the clock of the next ETC.

So how doe they actually know how to work and in what order ?. is their binary involved? Why is it that they just all increase by one once the input is high, How do the ICs know how to count up in order?

To answer your simple question simply, requires knowing how a 5 stage shift register works with negative binary feedback from the last stage meaning after reset to 00000 becomes 11111 after 5 clocks and 5 more gets back to ooooo. This gets your basic /10 counter to cascade the next digit stage.

...the other gates decode the 5 stage outputs into the combinations required to get the correct segments on for each of the 10 states. The design process uses a matrix of inputs and outputs using a Map with tricks enhanced by a method, "Karnaux map redundancy simplification" to find the minimum number of gates. After practise it gets easier.

Normally a 16 stage binary (successive /2) counter only needs 4 registers, but it is easier to decode 5 shift registers making sequential outputs than cascaded binary counts.

Look up Johnson Counter.

 

[djsfantasi]

I had a computer that was acting up. A DEC technician came in and adjusted the system clock by moving a blob of solder along two very long, spiraling copper traces.

Got a lot of "funny" ratings. What's funny is this is dead serious. Actually used to happen. Timings on the boards were set by adjusting the trace lengths with a solder bridge.

 

[Tony Stewart]

Maybe I am older than you guys. I learned logic with RTL, then DTL, then TTL, then LS-TTL then I threw away my TTL Cookbook and used Cmos then HC-Cmos for all logic circuits I designed.

I learned logic design from Philosophy 101 : Logic and Aristotle's Fallacies like Argumentum ad Hominem but wait there are 144 of these fallacies and our professor said you could find them every day in leading magazines and newspapers from false premises, invalid logic and thus false conclusions, not just different opinions.. But they only taught a dozen or so fallacies in 101. The others are pretty interesting especially if you synthesize them in hardware by accident.

The logic equations were the same that we learned in Digital Electronics, except they never taught fallacies in Engineering. It was always assumed our initial conditions were valid and race or metastable conditions was a post-grad subject. De Morgan et al based all their improvements in notation on Aristotle's logic.

These 3 basic logic operations OR, AND, NOT are made from three other fundamental components types; transistors, diodes and resistors. My 8 core i7 has a few billion transistors. Some of them may even open back doors to the NSA. But the elegance in these simple building blocks to make bigger ones reminds me of the DEMO CD for WordPerfect Six Point Oh that I got in the mail 2 decades ago which had songs that symbolized innovation in humanity as a segue to the world's 1st GUI word processor. ( pronounced seg'way) Example

My 3rd employer was still making core memory in Mexico, but I preferred CMOS, Schottky TTL for and differential emitter-coupled logic for 50, 10 and 2 ns speeds respectively.


My most favorite questions to test the skill of a Design Engineer's depth of experience, while recruiting was always to start with simple analog parameters for logic.
This is how fallacies are born and engineers learn about signal integrity.

Like ...
...can you float any TTL family input while breadboarding and what is the voltage at that unused input and what is it's logic level detected as?... then follow on what are the right and wrong ways to pull-up.

... and draw or describe 4 ways to debounce a switch.
= any wrong answer meant they hadn't done much breadboarding...

 

[spec]

I learned logic design from Philosophy 101 : Logic and Aristotle's Fallacies like Argumentum ad Hominem but wait there are 144 of these fallacies and our professor said you could find them every day in leading magazines and newspapers from false premises, invalid logic and thus false conclusions, not just different opinions.. But they only taught a dozen or so fallacies in 101. The others are pretty interesting especially if you synthesize them in hardware by accident.

Hy Tony,

You have touched on a big and fundamentally important subject here. It is because of beliefs in fallacies that many problems arise: global, local, macro, micro. I think that the main fallacies should be taught at a very early age in all schools.

spec

 

[Tony Stewart]

Hy Tony,

You have touched on a big and fundamentally important subject here. It is because of beliefs in fallacies that many problems arise: global, local, macro, micro. I think that the main fallacies should be taught at a very early age in all schools.

spec

My brother once wrote speeches for Cabinet Ministers and in big global newspapers. He 's really good at it for Political Science but he never learnt fallacies in PolySci and he uses them inadvertently to prove his points... ( fallaciously) like many global decisions.. as you agree.
The only place I know professionally that they learn some of the 144 fallacies is from Black's Rule of Law. you guessed it lawyers...go figure....Objection,,,,

 

[granddad]

I threw away my TTL Cookbook

Never had a TTL cookbook,,, had to manage with truth table pages in this...

[ my ETO portrait jpegs are never portrait ? ! ]

 

[spec]

grand, here is another TTL cookbook from the past: ftp://apollo.ssl.berkeley.edu/pub/cinema/04.%20Science/TTL%20Cookbook_0672210355.pdf

spec

 

[AnalogKid]

This link is on the desktop of all of my computers.

https://en.wikipedia.org/wiki/List_of_fallacies

ak